Approach to Neurologic Problems - Neuropathies

There is a local and perhaps nationwide shortage of neurologists, so what this means is that it takes a long time to get an appointment to see one. What that means is that primary care doctors are sometimes doing what they can to initially evaluate and sometimes manage neurologic problems. There are anyway a number of neurologic conditions that are quite common, things like headaches, neuropathy (or neuropathic symptoms), and weakness, as well as particular sorts of pain syndromes. I would also add that not all of these patients need to see a neurologist, mainly because there is little to do and management is often quite simple.

Let's start out in this series with the category of neuropathies, then focus on diabetic polyneuropathy. I would often see patients referred for electromyography (EMG) before they had seen a neurologist. A basic thing one could say about an EMG for neuropathy is that if you didn't know what was going on when you sent the patient for it, the test is unlikely to hand you the diagnosis on a platter. Given that situation, let's step back from this and go back to the patient.

Neuropathic symptoms
The two main functions of peripheral nerves are for sensation and movement, in particular muscular strength. Loss of sensory function would cause numbness or loss of feeling in the skin, but there are other kinds of sensation which for example have to do with feedback from joints to inform about arm, leg, and body position, and also feedback from muscles which allow for a sense of how much effort is being made, as well as the results of that effort.

Sensory nerve functions can be divided into two categories related to nerve fiber size. Very small and unmyelinated nerves carry signals related to pain and temperature. With loss of these functions, patients may say they injure themselves and don't realize it until they see they've cut themselves, or they have a hard time judging the temperature of bath water, especially with their feet. Myelinated nerve fibers carry signals related to some aspects of pain, but mainly light touch, vibration, and position sense. Loss in these areas may reduce fine motor manipulation such as is required for buttoning, or trying to pick out a particular object in a pocket based on feel. Loss of position sense impairs balance or coordinated activities, especially in the dark.

There is another kind of sensory nerve symptom which we can think of as either a nerve signal that gets messed up, so that normal stimulation is perceived as strange, maybe painful (called dysesthesias), or where there is a nerve sensation in the absence of stimulation, like pain or tingling (paresthesias).

Sensory nerve symptoms are quite common. We all have had at least brief episodes of tingling or numbness, maybe related to pressing on a nerve somewhere for too long.

Loss of motor function leads to loss of strength, or early on may mainly cause reduced stamina. Almost universally, patients will not see the difference. Strength has to do with the maximal force that can be generated from a group of muscles. Stamina has to do with how long you can apply that strength. There is another confounding issue which is important to consider, and this is sudden weakness which relates to pain from activity or effort.

Diabetic neuropathy as an example
This is probably the most common neuropathy most doctors will see. Diabetic patients are certainly aware of it, and at times I have had a hard time getting a patient to describe symptoms, since I keep getting responses like "I have [diabetic] neuropathy". What are your symptoms? "Neuropathy." Considering that diabetics can certainly have other kinds of neuropathy, this is important information. Start with the basic categories of location, kind of symptom, whether it occurs (mostly) at some time of day, any causative factors, any alleviating maneuvers.

It actually turns out that there are several different kinds of neuropathy. The most common is the symmetrical, distal neuropathy, where there is a progressive loss of nerve fibers, especially the small, unmyelinated fibers, so as expected there is loss of pain sensation, temperature sensation, and in some but not all there can be pain mostly or especially in the feet. Typically a burning character is described or acknowledged. I try to get patients to offer their own words for their descriptions before I suggest burning, sharp, dull, electric or some other type. There is a predilection for the burning to be worse or maybe only occur at night or bedtime. Often it is said to be especially bad after being on the feet or walking a lot during the day. Severe diabetic neuropathy may cause weakness, but much more often there is none or only loss of stamina.

Diabetics may also have focal neuropathies, often related to pressure. I consider diabetes a condition that sensitizes nerves to pressure, so you may see neuropathies at the elbow or knee (foot drop) more often in diabetics. Another much less common neuropathy is called diabetic amyotrophy. Although the name suggests only that there is muscle atrophy, it was shown some years ago that it is pathologically a kind of proximal nerve ischemia which produces most commonly proximal leg weakness, without prominent sensory symptoms. One of the happy days I had in practice was being able to "undiagnose" ALS (Lou Gehrig's disease) in a patient who actually had diabetic amyotrophy. At least in some cases, improving blood sugar control can result in much recovery.

The exam features that are important to document would be condition of the skin (evidence of trauma/ulcers), signs of muscular atrophy in some particular distribution, including the intrinsic foot and hand muscles. We're not expecting to see focal or asymmetric features in typical diabetic polyneuropathy. Much is made of loss of reflexes, though areflexia tends to be a weak element. With the sensory exam, you want to evaluate a mix of large fiber and small fiber functions. The former would be tested with light touch with some light stimulus (I used an aesthesiometer in practice, though in diabetics this can be difficult to interpret), and also proprioception -- how do you test that?

Proprioception involves in part the sense of where joints are or whether they have moved. The simple way to test is to hold a limb steady with one hand, then very slightly move a particular joint while the patient looks away. Do a series of repeated small motions, asking for a response of "up" or "down" with each movement. If you don't watch the movement yourself, you can compare your own sense and direction of movement as your benchmark. There is a learning curve for the patient, so expect to throw out initial errors as the process is demonstrated. Start with distal joints in fingers or toes, then as needed (if there are a lot of errors) work up to proximal joints of digits, then maybe even wrists or ankles (most of the time you won't get that far). Without labelling it as malingering, there is a certain instance of factitious reporting. Severe loss of proprioception (as high as the wrists or ankles) should impair function. This is the purpose of the Romberg test, where you have a patient stand with feet together, first with eyes open, then eyes closed. Impaired ankle proprioception will result in a positive Romberg with eyes closed (falling over with eyes open is something else completely). If you also have them hold their arms out, you watch for arms or hands drifting in unusual ways (drifting down and pronating the forearms only is a subtle sign of weakness, typically of CNS origin).

What about vibration? Yes you can test this, but here is a confession: I stopped assessing vibratory sense with regularity many years ago. Why? The biggest problem was that so many people have gradual loss of vibration sense as they get older, and the second biggest problem was that if that is the only thing on exam, it's almost worthless diagnostically. The standard way I was taught is also I think flawed. This was to smack the tuning fork on something, then hold the bottom end on a joint and have the patient tell you when the vibration goes away. There are uncertainties here. The first is that you know that by the time the patient indicates absence of vibration, some brief delay occurs from when the sensation stopped to when they spoke. Secondly, with a continuous vibration, you are likely to have persistence of perception at various levels in the nerve pathway. Here is a better method: on, off. Smack the tuning fork, then use an on, off technique asking the patient to indicate each time if they feel the vibration. As the vibration naturally decays you get a sense of amplitude from your own hand to compare. Don't forget the null stimulus -- some patients will say yes in the absence of vibration!

Small fiber function is mainly tested with sense of pin prick. The best instrument for this is a straight pin, and you only need to lightly touch the skin; no need to make a hole in it. Start distally and work your way up the arm or leg, perhaps 2 or 3 times to verify some level at which the pin becomes sharp, and in some way document that level for future reference. Even when it is felt sharp distally, there still may be a gradient of feeling as you come up the leg in particular. Some patients are hypersensitive distally. You can also test temperature sense, though I tend not to be too religious about this when the pin sense findings are sensible (sic).

The point of the motor exam, muscle strength testing, is to combine some measure of amount of strength with the overall distribution of any weakness. Years ago I used to see examiners referring to strength somewhere being "50%", but what does that mean? There is a tried and true method, the MRC scale, in which strength is assigned a number from 0 to 5, 0 being no ability to even tighten a muscle, 5 being normal strength. 3 means that the muscle can operate with full range of motion against gravity, so for the quadriceps/knee extension evaluation would be tested with patient seated. Any small addition of resistance by the examiner does not allow full range of motion. 4 means that some resistance can be overcome, but strength is less than normal, and 2 means there is some motion against gravity, but less than full range. Grade 1 is some tightening of the muscle without motion. Typically you want to see if there is a mainly proximal distribution of weakness, or distal, and furthermore, when weakness is distal, it tends to be worse in the ankles than wrists in neuropathies. A proximal pattern of weakness suggests something other than diabetic polyneuropathy, and is a hallmark of the various demyelinating neuropathies. There are patients who will not give a steady strong effort, or will suddenly give away. This cannot be graded, though sometimes a brief strong effort is long enough to use for reasonably good grading of strength.

Something worth saying about this exam is that you get better at it the more times you do it. To develop and maintain a sense of what is normal for different ages, you should do parts of this in patients without neuropathy symptoms.

In summary of the history and exam features of diabetic polyneuropathy, if you see a diabetic patient with loss of pain sense and perhaps painfulness at nighttime, associated with some loss of pin sensation on exam, and preservation of light touch, proprioception (position sense), and little or no weakness, they have diabetic polyneuropathy. An EMG is not going to add anything, and typically for my own patients fitting this picture I don't even order an EMG.

Curiously, there are nondiabetics who also have this same picture. Some of them will develop diabetes later, but others never do. The approach is no different.

Management and some curious features
A newly diagnosed diabetic of course needs that managed as appropriate. Curiously, there doesn't seem to be any good correlation between the severity or duration of the diabetes and the severity of the associated neuropathy. There is not always a correlation between degree of control of sugars and the likelihood or severity of the neuropathy. One paradoxical thing I have seen is a worsening of neuropathic symptoms with better control of sugars. This is typically temporary and occurs when sugars have been severely out of control. I think this is because severe hyperglycemia has some anesthetic effect on nerve function. Even though there is this discrepancy of severity of diabetes and severity of neuropathy, I still believe that for the individual patient, they will do better with better sugar control, not the least because of the other various complications of diabetes.

I'm not going to discuss management of painful neuropathy, but save this for the next post. One thing to say in summary so far is that for the typical diabetic, there isn't such a great need for a neurologist, at least to make the diagnosis. One can save time and money by not getting EMGs routinely on these patients.